Sparse Singular Value Decomposition with Missing Data

• Activity Number: 299
• Type: Topic Contributed
• Date/Time: Tuesday, August 6, 2013 : 8:30 AM to 10:20 AM
• Sponsor: Section on Statistical Learning and Data Mining
• Author(s): Tingni Sun*+ and Zongming Ma
• Companies: University of Pennsylvania and University of Pennsylvania
• Keywords: Singular value decomposition ; sparsity ; missing data

Abstract

In this talk, we focus on sparse singular value decomposition in the high-dimensional setting with missing observations. We propose a data-driven procedure to handle missing data and estimate the subspaces spanned by leading left and right singular vectors and also the matrix itself. Theoretical results are provided under proper sparsity condition on the true singular vectors. The Bernoulli model is used to describe the missing scheme.